基于变分模态分解与独立分量分析的轴承故障特征提取方法

发布时间：2018-01-20 03:35

本文关键词： 滚动轴承变分模态分解独立分量分析故障诊断　出处：《济南大学学报(自然科学版)》2017年03期 　论文类型：期刊论文

【摘要】：针对滚动轴承早期故障振动信号能量小且易受背景噪声干扰,从而导致故障特征提取困难等问题,提出基于变分模态分解(VMD)与独立分量分析(ICA)相结合的故障特征提取方法;该方法首先将原始故障信号进行VMD,得到若干正交的本征模态分量(IMF),然后依据峭度准则对分解后的信号进行分组重构,作为ICA的输入矩阵,最后采用Fast ICA算法实现故障信号与噪声信号的分离,从而提取机械故障特征信息;将轴承故障数据作为研究对象进行故障特征提取,并与集成经验模态分解-独立分量分析(EEMD-ICA)方法对特征信号的提取效果进行对比。结果表明,基于VMD与ICA的轴承故障特征提取方法提高了分解效率,解决了信号易受噪声干扰的问题,实现了轴承故障的精确诊断。
[Abstract]:Aiming at the problem that the vibration signal of early fault of rolling bearing is small in energy and easily disturbed by background noise, it is difficult to extract fault features. A fault feature extraction method based on variational mode decomposition (VMD) and independent component analysis (ICA) is proposed. In this method, the original fault signal is first processed into ICA, and some orthogonal eigenmode components are obtained. Then, the decomposed signal is grouped and reconstructed according to the kurtosis criterion as the input matrix of ICA. Finally, the Fast ICA algorithm is used to separate the fault signal from the noise signal, so as to extract the mechanical fault feature information. The bearing fault data is used as the research object to extract the fault features. And compared with the integrated empirical mode decomposition-independent component analysis (EEMD-ICA) method for feature signal extraction, the results show that. The method of bearing fault feature extraction based on VMD and ICA improves the decomposition efficiency, solves the problem that the signal is vulnerable to noise interference, and realizes the accurate diagnosis of bearing fault.
【作者单位】：石家庄铁道大学电气与电子工程学院;
【基金】：国家自然科学基金项目(11227201,11372199,11572206) 河北省自然科学基金项目(A2014210142)
【分类号】：TH133.3
【正文快照】： 利用振动信号进行分析是设备故障诊断中最直接有效的方法。在机械系统中,由于受复杂背景噪声以及其他干扰源的影响,早期故障的特征信息往往被淹没在噪声中,导致其低频段特征频率与谐波特征频率极其微弱,因此故障特征很难提取出来。有效地提取故障信号一直是机械设备监测和故障

【相似文献】